The p53 tumor suppressor initiates cell cycle arrest, senescence, and apoptosis programs to prevent abnormal cell proliferation. All three activities contribute to tumor suppression in vivo. We and others have performed wild-type p53 reconstitution experiments in tumors that lacked p53 and observed tumor regression. Deletions of p53 are, however, uncommon in human cancers. In fact, more than 80% of tumors with alterations of the p53 gene acquire missense mutations in p53 rather than p53 deletions. This distinction is critical because p53 missense mutations clearly display gain-of-function and/or dominant-negative activities in vivo. Additionally, over-expression of MDM2 and MDM4 which encode p53 inhibitors are also common events in tumorigenesis and contribute to inactivation of the p53 pathway. The objective of this proposal is to genetically reconstitute wild-type p53 in the context of a tumor cell with the kinds of alterations that are observed in human cancers. Intriguing preliminary data from my laboratory indicates that tumors that develop as a result of a p53 mutation rather than loss of p53 respond differently to reactivation of wild-type p53: tumor responses in the p53 mutant background are muted and lead to tumor stasis, not tumor regression, by unknown mechanisms. Aim 1 is focused on unraveling these mechanisms. Although p53 has not yet been reactivated in tumors with high levels of Mdm2 or Mdm4, we surmise from their exquisite regulation of p53 that high levels of Mdm2 and Mdm4 would swamp out the activities of a reintroduced p53. We plan to increase wild-type p53 levels in tumors with high levels of Mdm2 and Mdm4 (aim 2). Lastly, data suggest that the genotype of the cells surrounding the tumor affects tumor growth and therapeutic response. Current mouse models do not address the role of the surrounding normal tissue as all cells of the genetically modified mouse have p53 defects. We plan to create and evaluate a somatic model that recapitulates a p53 mutation in a single cell surrounded by normal cells (aim 3).